ORCHIDEE-MICT (v8.4.1), a land surface model for the high latitudes: model description and validation

HAL Id: hal-01806766

https://hal.archives-ouvertes.fr/hal-01806766

Submitted on 8 Sep 2020

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

the high latitudes: model description and validation

Matthieu Guimberteau, Dan Zhu, Fabienne Maignan, Ye Huang, Chao Yue, Sarah Dantec-Nédélec, Catherine Ottlé, Albert Jornet-Puig, Ana Bastos,

Pierre Laurent, et al.

To cite this version:

Matthieu Guimberteau, Dan Zhu, Fabienne Maignan, Ye Huang, Chao Yue, et al.. ORCHIDEE-MICT

(v8.4.1), a land surface model for the high latitudes: model description and validation. Geoscientific

Model Development, European Geosciences Union, 2018, 11 (1), pp.121 - 163. �10.5194/gmd-11-121-

2018�. �hal-01806766�

Geosci. Model Dev., 11, 121–163, 2018 https://doi.org/10.5194/gmd-11-121-2018

© Author(s) 2018. This work is distributed under the Creative Commons Attribution 3.0 License.

ORCHIDEE-MICT (v8.4.1), a land surface model for the high latitudes: model description and validation

Matthieu Guimberteau

, Dan Zhu

, Fabienne Maignan

, Ye Huang

, Chao Yue

, Sarah Dantec-Nédélec

,

Catherine Ottlé

, Albert Jornet-Puig

, Ana Bastos

, Pierre Laurent

, Daniel Goll

, Simon Bowring

, Jinfeng Chang

, Bertrand Guenet

, Marwa Tifafi

, Shushi Peng

, Gerhard Krinner

, Agnès Ducharne

, Fuxing Wang

, Tao Wang

, Xuhui Wang

, Yilong Wang

, Zun Yin

, Ronny Lauerwald

, Emilie Joetzjer

, Chunjing Qiu

,

Hyungjun Kim

, and Philippe Ciais

1

Laboratoire des Sciences du Climat et de l’Environnement, LSCE/IPSL, CEA – CNRS – UVSQ, Université Paris-Saclay, 91191 Gif-sur-Yvette, France

2

Sorbonne Universités (UPMC), CNRS-IRD-MNHN, LOCEAN/IPSL, 4 place Jussieu, 75005 Paris, France

3

Sino–French Institute for Earth System Science, College of Urban and Environmental Sciences, Peking University, Beijing 100871, China

4

CNRS, Univ. Grenoble Alpes, Institut des Géosciences de l’Environnement (IGE), 38000 Grenoble, France

5

UMR 7619 METIS, Sorbonne Universités, UPMC, CNRS, EPHE, 4 place Jussieu, 75005 Paris, France

6

Laboratoire de Météorologie Dynamique, Ecole Polytechnique, 91128 Palaiseau, France

7

Key Laboratory of Alpine Ecology and Biodiversity, Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing 100085, China

8

CAS Center for Excellence in Tibetan Plateau Earth Sciences, Chinese Academy of Sciences, Beijing 100085, China

9

Laboratoire de Météorologie Dynamique, Université Pierre et Marie Curie, 75005 Paris, France

10

Université Libre de Bruxelles, Belgium

11

University of Exeter, Exeter, UK

12

CNRS, Université Paul Sabatier, ENFA; UMR5174 EDB (Laboratoire Evolution et Diversité Biologique), 118 route de Narbonne, 31062 Toulouse, France

13

Institute of Industrial Science, The University of Tokyo, Tokyo, Japan

*

These authors contributed equally to this work.

Correspondence: Matthieu Guimberteau (matthieu.guimberteau@lsce.ipsl.fr) and Dan Zhu (dan.zhu@lsce.ipsl.fr) Received: 17 May 2017 – Discussion started: 16 June 2017

Revised: 4 October 2017 – Accepted: 27 November 2017 – Published: 15 January 2018

Abstract. The high-latitude regions of the Northern Hemi- sphere are a nexus for the interaction between land surface physical properties and their exchange of carbon and energy with the atmosphere. At these latitudes, two carbon pools of planetary significance – those of the permanently frozen soils (permafrost), and of the great expanse of boreal forest – are vulnerable to destabilization in the face of currently ob- served climatic warming, the speed and intensity of which are expected to increase with time. Improved projections of future Arctic and boreal ecosystem transformation require improved land surface models that integrate processes spe- cific to these cold biomes. To this end, this study lays out rel-

evant new parameterizations in the ORCHIDEE-MICT land

surface model. These describe the interactions between soil

carbon, soil temperature and hydrology, and their resulting

feedbacks on water and CO

fluxes, in addition to a re-

cently developed fire module. Outputs from ORCHIDEE-

MICT, when forced by two climate input datasets, are ex-

tensively evaluated against (i) temperature gradients between

the atmosphere and deep soils, (ii) the hydrological com-

ponents comprising the water balance of the largest high-

latitude basins, and (iii) CO

flux and carbon stock observa-

tions. The model performance is good with respect to empir-

ical data, despite a simulated excessive plant water stress and

a positive land surface temperature bias. In addition, acute model sensitivity to the choice of input forcing data suggests that the calibration of model parameters is strongly forcing- dependent. Overall, we suggest that this new model design is at the forefront of current efforts to reliably estimate future perturbations to the high-latitude terrestrial environment.

1 Introduction

At the high latitudes, the complex coupling between soil ther- mal and hydraulic processes, snowpack properties, and plant and soil carbon pools is of great importance. Snow accumu- lation and freezing of soil water lead to a net storage of water from October to April. Through the processes of snowmelt and the onset of soil thaw in spring, water is made avail- able for plant uptake and growth. Simultaneously, however, much of this is “lost” as runoff to rivers, causing peak dis- charge rates in May–June (Yang et al., 2003) and the flood- ing of large flatland areas from May to September (Papa et al., 2008; Biancamaria et al., 2009). In summertime, the peak in incident solar radiation causes a temperature max- imum that increases water evaporative demand on the land surface. Many boreal and Arctic regions thus have a nega- tive water balance in summer (Schulze et al., 1999), which may impose powerful constraints on plant growth. Siberian and Canadian boreal forests have thus been shown to expe- rience water stress, with ratios of surface sensible to latent heat flux of up to ∼ 2 (Jarvis et al., 1997; Baldocchi et al., 1997; Schulze et al., 1999) causing further heating of the near-surface atmosphere.

These large seasonal shifts of the high-latitude water bal- ance – how water input from precipitation is shared between changes in water storage in snow, ice and soil moisture, and balanced against losses from evapotranspiration, sublimation and river discharge – can now be better assessed and eval- uated using state-of-the-art observation datasets. In addition, in terms of realistic process representation, land surface mod- els (LSMs) focusing on high-latitude phenomena require the inclusion of the following non-exhaustive series of pivotal hydrological and biogeochemical interactions.

1. A representation of permafrost physics and seasonal freeze–thaw cycles, which determine the soil hydrologic and thermal budgets and the volume and timing of lat- eral water flows to rivers.

2. The impact of winter snow acting as an insulating “bar- rier” between soils and overlying air from fall to early spring. These have subsequent effects on soil tempera- ture and water content, feeding back onto snow thick- ness itself.

3. The seasonal mediation of plant water availability via snowmelt water, transpiration losses and the depth of the permafrost table (active layer thickness), which in

turn determine the availability of the lateral water flows that feed rivers in the warmer months.

4. The limitations on plant productivity and biomass due to acute climatic conditions in high-latitude regions. These primarily involve biotically prohibitive cold tempera- tures from fall to late spring, low soil moisture in dry- summer regions, and fire events caused by hot and dry conditions.

5. The buildup of large soil carbon stocks under cold con- ditions through the slow burial of organic matter in the permafrost via cryoturbation and sedimentary soil for- mation processes (e.g., Hugelius et al., 2013; Tarnocai et al., 2009).

6. Feedbacks between high soil carbon concentrations and profiles of soil temperature, water and permafrost car- bon content (e.g., Lawrence and Slater, 2008; Decharme et al., 2016).

We represent the above processes in an updated version of the ORCHIDEE LSM (ORganizing Carbon and Hydrol- ogy in Dynamic EcosystEms), known as ORCHIDEE-MICT (aMeliorated Interactions between Carbon and Temperature), which we describe in this study. Since the comprehensive de- scription of the ORCHIDEE model by Krinner et al. (2005), the model has gone through major modifications and im- provements; we present here the major ones linked to high- latitude processes. ORCHIDEE-MICT is evaluated over the last 2 to 3 decades (depending on the variable) against empir- ically generated datasets. Against these, we are able to eval- uate model performance regarding the distribution of per- mafrost and the effect of snow on soil thermics (mecha- nisms 1 and 2); the different components of the water cy- cle over a wide range of high-latitude basins (mechanism 3);

plant primary productivity as constrained by high-latitude conditions (mechanisms 3 and 4); and replication of soil car- bon stocks and feedback dynamics (mechanisms 5 and 6).

2 ORCHIDEE model overview

The starting point for our updated land surface model

(ORCHIDEE-MICT) is ORCHIDEE-TRUNK revision

3976. As detailed in Sect. 3, its description of soil tem-

perature and vertical water transport dynamics is based

on coupled diffusion equations with identical vertical dis-

cretization (F. Wang et al., 2016), and includes soil freezing,

its effect on water infiltration, and phase change-induced

heat sources and sinks in the soil column (Gouttevin et al.,

2012a). The snow model described by Wang et al. (2013) is

incorporated into this version, where snow is discretized into

three layers of variable thickness, conductivity and density,

accounting for snow liquid water content (Boone and

Etchevers, 2001). In terms of large-scale hydrology, a river

routing scheme including floodplains and their dynamics

M. Guimberteau et al.: ORCHIDEE-MICT, a LSM for the high latitudes 123 (d’Orgeval et al., 2008; Guimberteau et al., 2012) is coupled

to simulated grid-cell runoff (Sect. 3.2), permitting the calculation of “natural” river discharge (i.e., in the absence of dams or human water withdrawals).

The carbon cycle model includes half-hourly photosyn- thesis (GPP), daily allocation of GPP assimilates to au- totrophic respiration and eight plant biomass pools (foliage, roots, above-/below-ground sapwood and heartwood, fruits and carbon reserves), and prognostic phenology (Botta et al., 2000). These pools are characterized by different turnover times, mortality rates and subsequent litter and soil carbon decomposition rates. Litter carbon is funneled between struc- tural and metabolic fractions, and soil carbon between active, slow and passive pools, following Parton et al. (1987).

The model divides vegetation into 13 plant functional types (PFTs). Each PFT follows the same suite of equations but with PFT-specific parameter values and phenology func- tions (Krinner et al., 2005). PFT fractions are assigned to three soil tiles corresponding to bare soil, short vegetation (grass and crop PFTs) and forests (all tree PFTs). The soil moisture budget of each soil tile is calculated separately, but different PFTs in the same soil tile interact as they share the same soil moisture source. While transpiration is calculated separately for each PFT, and soil moisture for each soil tile, the energy budget of a grid cell with multiple PFTs is calcu- lated using the area-weighted average of those PFTs. This in turn defines mean grid-cell land surface temperature, giving the upper boundary condition for the vertically discretized soil thermal module.

Temperature, water and carbon interactions described in ORCHIDEE revision 3976 are summarized in Fig. 1 by the black arrows. Air temperature and humidity impact phenol- ogy, photosynthesis, autotrophic respiration and the water and heat fluxes comprising the surface energy budget. Soil moisture in the root zone modulates photosynthesis and tran- spiration, which depends on wilting point and field capacity.

In ORCHIDEE revision 3976, while soil carbon decompo- sition is impacted by soil water and temperature, soil car- bon stocks themselves exert no feedback on the soil physical state.

The key model developments presented here in ORCHIDEE-MICT (v8.4.1) thus include the feedback effects of soil organic carbon (SOC) concentration on both soil thermic and soil water dynamics (Fig. 1, red arrows).

Because these SOC-affected soil physics alter the above- and below-ground components of the carbon cycle, as well as plant transpiration via hydraulic stress, we can expect com- plex indirect effects on the energy, water and carbon budgets (Fig. 1). Note that in the simulations here, soil thermal and hydrological modules read a prescribed observational SOC map (NCSCD in permafrost regions and HWSD in non- permafrost regions) instead of the prognostically simulated SOC, to exclude the impact of bias in the carbon cycle module, for the purpose of model evaluation for the present day in this study. Note that several other updates were

implemented in ORCHIDEE-TRUNK (revision 3976) and passed to ORCHIDEE-MICT (v8.4.1), including a revised background albedo based on satellite observations, and up- dates of the photosynthesis scheme. These will be described in an upcoming paper for ORCHIDEE-TRUNK (version close to revision 3976) that will be used for the CMIP6 exercise. In the following, we describe the parameterizations that define ORCHIDEE-MICT (v8.4.1).

3 High-latitude processes in the initial ORCHIDEE version

3.1 Soil freezing and snow processes

The soil freezing scheme developed by Gouttevin et al.

(2012a) describes phase changes of soil water, simulating the latent heat exchanges involved in the freezing and melting of soil water, and subsequent changes in thermal and hydrologi- cal ground properties. Soil heat conductivity and heat capac- ity are dependent on soil ice content. The hydraulic conduc- tivity of the soil is parameterized according to its liquid wa- ter content and decreases with the frozen soil fraction. Heat transfer through the soil column is represented by a one- dimensional heat conduction equation, with latent heat act- ing as a source or sink term (Gouttevin et al., 2012a), in the following function:

c ∂T

∂t = ∂

∂z

λ ∂T

∂z

+ ρ

L ∂θ

∂t , (1)

where c is volumetric soil heat capacity (J K

m

); T is soil temperature (K); λ is soil thermal conductivity (J m

s

K

); ρ

is ice density (kg m

); L is latent heat of fusion (J kg

); θ

is volumetric ice content (m

m

);

t is time (s) and z is depth (m). In ORCHIDEE-MICT, this equation is discretized on the 32 vertical layers of the model with a total soil depth of 38 m (Fig. S1 in the Supplement).

Note that the soil hydrology has only 11 layers up to 2 m, so the volumetric contents of water and ice below 2 m take the values of the bottom layer (i.e., the 11th layer).

Snowpack is represented by a three-layer snow model of intermediate complexity, as described in Wang et al. (2013).

This scheme was implemented to resolve the energy and wa-

ter budgets inside the snowpack, accounting for thawing and

refreezing of liquid water. The snow model produces prog-

nostic snow temperature, density and SWE for the three snow

layers. Modifications were recently implemented to represent

snowpack sub-grid-scale variability. A snow cover fraction

over the grid cell was introduced as a function of SWE. This

was used for improving albedo and surface temperature es-

timates. Although this snow cover fraction was calculated

for glacier and vegetated-surface areas separately, it is not

dependent on the vegetation cover. Additional modifications

were implemented (uniformization of the energy budget cal-

culation, update of the snow-covered vegetation albedo. . . )

Figure 1.

Temperature, water and carbon interactions in the initial version of ORCHIDEE (black), and processes included in ORCHIDEE- MICT in this study (red). Note that in the simulations in this study, the soil thermics and hydrology modules read a prescribed observation- based soil carbon map (see Eq. 9), which is independent of the prognostically simulated SOC by the carbon module; thus, the two red arrows here are dashed lines.

and will be described in the upcoming CMIP6 ORCHIDEE paper as mentioned in Sect. 2.

3.2 Soil hydrology and river routing

ORCHIDEE simulates soil water fluxes and storage through a multi-layer soil hydrology scheme described by de Rosnay et al. (2000, 2002) and Campoy et al. (2013). Soil moisture is redistributed in the column by solving the Richards equa- tion for vertical unsaturated flow under the effect of root up- take. The hydraulic conductivity and diffusivity depend on soil moisture, following the Mualem–van Genuchten model (Mualem, 1976); (Van Genuchten, 1980), and using param- eters defined by Carsel and Parrish (1988). These variables depend on the dominant soil texture in each grid cell, based on the 12 USDA texture classes provided at the 0.08

reso- lution from Reynolds et al. (2000). For frozen soils, the de- crease in the hydraulic conductivity (Gouttevin et al., 2012a) reduces infiltration into the soil and drainage, and increases surface runoff. The 2 m soil column is divided into 11 lay- ers, with layer thickness increasing geometrically with depth (Fig. S1). The saturated hydraulic conductivity is modified according to the scheme in d’Orgeval et al. (2008). This decreases exponentially below a top-30 cm depth boundary to account for increased soil compaction, as suggested by Beven and Germann (1982), and increases above that bound- ary towards the soil surface due to the enhanced infiltra- tion capacity afforded by vegetative roots, whose presence increases soil porosity in the root zone (Beven, 1984). The canopy throughfall rate and soil hydraulic conductivity gov- ern the partitioning between surface runoff and soil infiltra-

tion. This partitioning involves a time-splitting procedure in- spired by Green and Ampt (1911), describing the propaga- tion of the wetting front. The second physical factor con- tributing to total runoff is free gravitational drainage at the bottom of the soil.

The runoff routing module (Polcher, 2003; Ngo-Duc et al., 2005; Guimberteau et al., 2012) aggregates surface runoff and drainage produced at a 30 min time step to calculate daily flow between grid cells and discharge to the ocean.

Grid cells are subdivided into basins in which water is trans- ferred through a series of linear reservoirs along the drainage network, derived from a 0.5

resolution dataset (Vörösmarty et al., 2000; Oki et al., 1999). In a given basin, a “slow” reser- voir collects drainage water, while a “fast” reservoir collects surface runoff, each with different linear response timescales.

Corresponding outflows are transferred to the stream reser- voir of the downstream basin. The process is fully detailed in Guimberteau et al. (2012).

The routing scheme includes a parameterization of flood-

plains (d’Orgeval et al., 2008; Guimberteau et al., 2012),

the maximum extent of which is prescribed by the GLWD

(Global Lakes and Wetlands Database) map (Lehner and

Döll, 2004). In grid cells with flooded areas, river discharge

from upstream basins is diverted to a floodplain reservoir,

which then feeds a delayed return flow back to the stream

reservoir of the basin.

M. Guimberteau et al.: ORCHIDEE-MICT, a LSM for the high latitudes 125 4 New processes and parameterizations

4.1 Soil carbon discretization

In ORCHIDEE-MICT, the three soil carbon pools (active, slow and passive) share a common 32-layer discretization scheme with that of soil temperature, to a maximum depth of 38 m. Carbon inflows to the soil pools from decomposed litter are partitioned along this depth using an exponential function that corresponds to the prescribed PFT root profile.

Decomposition of soil carbon is calculated at each layer as a function of soil temperature, moisture, and texture (Koven et al., 2009; Zhu et al., 2016). Vertical mixing of soil car- bon due to cryoturbation (mixing of soil layers induced by repeated freeze–thaw cycles) and bioturbation are accounted for by adding a diffusion term in the soil carbon equation:

∂C

(z, t )

∂t = I

(z, t ) − g

(z, t ) C

(z, t ) + D ∂C

(z, t )

∂z

, (2) where C

(z, t ) is carbon content of pool i at depth z and time t (g C m

); I

(z, t ) is carbon input (g C m

d

);

g

(z, t ) is decomposition rate (d

); D is diffusive mixing rate, set as 10

m

yr

through the active layer, and de- creases linearly to zero at 3 m in permafrost regions, to rep- resent cryoturbative mixing (Koven et al., 2009), and set as 10

m

yr

above 2 m in non-permafrost regions to repre- sent bioturbation (Koven et al., 2013).

4.2 SOM-dependent soil thermal and hydraulic parameters

Soil organic matter (SOM) significantly modifies soil ther- mal and hydraulic properties. SOM lowers thermal conduc- tivity and increases heat capacity (e.g., Lawrence and Slater, 2008; Decharme et al., 2016), and increases soil porosity, which in turn increase saturated hydraulic conductivity and available water capacity (e.g., Hudson, 1994; Morris et al., 2015). As a consequence, the presence of SOM modulates heat transfer from the surface through the soil column, typ- ically leading to cooler soil temperature during summer.

SOM-effected increases in soil water holding capacity also enhance plant available water and thus primary productiv- ity (Krull et al., 2004) and transpiration. SOM impacts on soil thermics and hydraulics have previously been parameter- ized in the global LSMs CLM (Lawrence and Slater, 2008), JULES (Chadburn et al., 2015b) and ISBA (Decharme et al., 2016). In ORCHIDEE, SOM thermal insulation was previ- ously investigated by Koven et al. (2009), but its parameter- ization was imbedded in a prior model version which em- ployed bucket-type soil hydrology. This, however, is not ap- plicable to ORCHIDEE-MICT, which uses a new vertically discretized hydrology scheme and its coupling with the ther- mal module. In addition, the Koven et al. (2009) study did not include SOM effects on soil hydraulic properties, which

are addressed in ORCHIDEE-MICT and described in detail below.

Thermal conductivity and heat capacity

By default, soil thermal conductivity and heat capacity in ORCHIDEE are calculated in each soil layer as empirical functions of the 12 USDA soil texture classifications (see Ta- ble S1 in the Supplement) and soil water and ice contents, following F. Wang et al. (2016):

λ

= Ke

λ

+ (1 − Ke

)λ

, (3)

with

λ

= λ (

)

i,solid

λ

θi,sat θi,liq θi,liq+θi,ice

liq

λ

θi,sat θi,ice θi,liq+θi,ice

ice

, (4)

c

= c

+ θ

c

+ θ

c

, (5)

where λ

and λ

are saturated and dry thermal con- ductivities for layer i; λ

and λ

are thermal conductivi- ties of liquid water and ice, equaling 0.57 and 2.2, respec- tively (W m

K

); λ

is thermal conductivity of soil solids (see Table S1); c

and c

are heat capacities of liq- uid water and ice, equaling 4.18 10

and 2.11 10

, respec- tively (J K

m

); c

is dry soil heat capacity depending on soil texture; θ

is volumetric moisture content at sat- uration (porosity), and it varies with soil textures; θ

and

θ

are prognostic volumetric liquid water and ice contents

(m

m

) that are computed by the soil hydrology model; Ke

is the Kersten number given by the following.

For unfrozen soils:

Ke

=

( log

(S

) + 1 0.7 log

(S

) + 1

0

if

( S

> 0.1 0.05 < S

≤ 0.1

S

≤ 0.05

(6) with

S

= θ

θ

. (7)

For frozen soils:

Ke

= S

, (8)

where S

is the degree of saturation.

To account for the impacts of organic carbon on soil ther- mal properties in ORCHIDEE-MICT, we follow Lawrence and Slater (2008) in assuming that soil physical properties are weighted averages of mineral soil (as the default values in standard ORCHIDEE) and pure organic soil, with the or- ganic soil fraction f

calculated as

f

= min

1, ρ

ρ

, (9)

where ρ

is the carbon content for layer i (kg C m

),

derived from an observation-based soil organic carbon map

from NCSCD (Hugelius et al., 2013) in permafrost regions and from HWSD (FAO, 2012) in non-permafrost regions, af- ter linear vertical interpolation from their original soil hori- zons to fit ORCHIDEE-MICT vertical layers; ρ

equals 130 kg C m

, a typical soil carbon density of peat (Lawrence and Slater, 2008). Therefore, the parameters in Eqs. (3)–(7) are calculated as

P

= (1 − f

) P

+ f

P

, (10)

where P

represents different properties λ

, λ

, c

, and θ

. The values of P

for each soil texture and P

are listed in Table S1. Note that here we followed Lawrence and Slater (2008) to use linear weighting organic and min- eral soil properties, while in some other models like JULES (Chadburn et al., 2015a) and ISBA (Decharme et al., 2016), soil thermal conductivities are calculated as geometric aver- ages of organic and mineral soils, consistent with the treat- ment for soil water and ice (Eq. 4). The geometric averaging method increases the effect of the organic fraction compared to arithmetic averages, and would be tested in ORCHIDEE- MICT in future developments.

Available water capacity

Plant available water capacity, defined as the difference in the amount of water held by each soil layer between field capac- ity (θ

) and permanent wilting point (θ

), determines the capacity of the soil to store and supply water for plants, and is therefore an important aspect of soil fertility (Hudson, 1994).

For mineral soils in ORCHIDEE, θ

and θ

are derived from measurements of the soil matric potential at field capac- ity and wilting point, based on the soil water retention curve described by the van Genuchten equation (Van Genuchten, 1980):

θ = (θ

− θ

) 1 + (α ( − 9))

1−¹_n

+ θ

, (11)

where ψ is soil matric potential (kPa), and ψ = − 33 kPa (or

− 10 kPa for the three sandy soils; see Table S2) corresponds to field capacity θ

, while ψ = − 1500 kPa corresponds to wilting point θ

for all textures; θ

is the residual volumetric water content (m

m

); α and n are empirical fitting coef- ficients, with their values for different soil textures listed in Table S2.

SOC has been shown to significantly increase water re- tention (Rawls et al., 2003). To parameterize this SOM ef- fect, we assume that θ

and the coefficients in Eq. (11) do not change with carbon content, while porosity θ

increases with organic carbon (Eq. 10). Therefore, both θ

and θ

in- crease under higher carbon contents, but θ

increases faster, resulting in a higher available water capacity (Fig. S2), con- sistent with the patterns observed in Hudson (1994).

4.3 Reformulation of soil hydric stress above the permafrost table

It is known that reduced soil moisture availability decreases the rate of photosynthesis, but the parameterization of this photosynthetic stress differs amongst models (Medlyn et al., 2015). ORCHIDEE-MICT lacks a fully mechanistic plant hydraulic structure that calculates plant internal water move- ment via constraints from water potential (ψ) and conduc- tance of roots, stems and leaves. Instead, a stress factor, which ranges from 0 to 1, is calculated based on the rela- tive moisture content at each soil layer. This factor is applied to stomatal conductance and mesophyll conductance, as well as the maximum RuBisCO activity rate (Vcmax) and maxi- mum electron transport rate (Jmax), in order to account for experimentally observed effects of drought on stomatal and non-stomatal photosynthetic limitation (Zhou et al., 2014).

The stress factor (γ ) of water limitation is calculated as γ

= θ

− θ

θ

+ ρ(θ

− θ

) , (12)

γ =

11

X

i=1

γ

w

, (13)

where γ

is relative moisture content at each soil layer i, bounded between 0 and 1; ρ represents the fraction above which photosynthesis rate is not limited by soil moisture, and is set at 0.8; w

is the weighting factor for each layer.

In the initial version of ORCHIDEE, the profile of w

was assumed to be constant over time, although it differed be- tween tree and grass PFTs, with the highest value at 1.5 m depth for trees and 0.37 m depth for grasses. We considered this description inappropriate for the high latitudes, and in particular for permafrost regions, where trees develop shal- low and lateral roots above the permanently frozen layer (Ka- jimoto et al., 2003). Thus, in ORCHIDEE-MICT, w

is mod- ified to be a dynamic profile which optimizes plant water use, in a manner inspired by the representation given in Beer et al.

(2007):

w

= γ

P

i=1

γ

, (14)

where if layer i is below the modeled active layer thickness, w

is set to zero, and the remaining w are re-normalized to one.

4.4 Fires

The SPITFIRE (SPread and InTensity of FIRE) prog- nostic fire module, which has been previously integrated into and calibrated for a standard version of ORCHIDEE (Thonicke et al., 2010; Yue et al., 2014), was merged into ORCHIDEE-MICT. Ignitions were re-calibrated us- ing the GFED4s dataset (http://www.globalfiredata.org/data.

html) by using region-specific scaling factors (see Table S3)

M. Guimberteau et al.: ORCHIDEE-MICT, a LSM for the high latitudes 127 and the exclusion of cropland fires to ensure that simulated

mean annual burned area for 1997–2013 was equal to that of the GFED4s dataset. Note that this method only calibrated for mean annual regional burned area, and that simulated lat- itudinal distributions and grid cell spatial patterns of burned area and fire carbon emissions, and their interannual and sea- sonal variabilities, could still be compared with observation- based data. Deforestation and peatland fires are not explic- itly simulated, but as both fire types rely on suitable weather conditions to occur, which could be partly captured by SPIT- FIRE (Yue et al., 2015), model simulations are expected to partially include these fire types.

5 Simulation protocol, forcing and evaluation datasets 5.1 Simulation protocol and forcings

5.1.1 Simulation setup

Two separate runs using different climate forcing input data – CRUNCEP v7 (hereafter CRUNCEP) and GSWP3 – were performed with ORCHIDEE-MICT for the terrestrial North- ern Hemisphere (> 30

N) at 1

spatial resolution. Both sets of runs encompass the 20th century and the beginning of the 21st century, and were preceded by separate spin-ups for each climate dataset, forced by fixed pre-industrial condi- tions of atmospheric CO

(286 ppm) and vegetation maps.

The dynamic vegetation model is de-activated throughout both runs. In order to accumulate soil organic carbon in the model, which requires substantial computing time be- fore reaching near-equilibrium in the presence of the slow mixing processes described in Sect. 4.1, the spin-up proce- dure comprised three steps. (1) The full ORCHIDEE-MICT model was forced by looped climate fields over the period 1960–1990 for 100 years to reach equilibria for soil tempera- ture, soil moisture, vegetation productivity, soil carbon inputs from dead plants, etc. We used the 1960–1990 loop, instead of pre-industrial climate, to approximate the higher Holocene temperatures relative to the “pre-industrial” period that have been reconstructed in Marcott et al. (2013). (2) A soil car- bon sub-model was run for 20 000 years, forced by the out- puts from the preceding step. (3) The full ORCHIDEE-MICT model was run for 100 years, forced by looped 1901–1920 climate data, to approach to the pre-industrial equilibrium for physical variables, carbon fluxes, and fast carbon pools. A fi- nal transient simulation from 1861 to 2007 (using the 1901–

1920 climate loop for the period 1861–1900 due to the lack of climate forcing before the 20th century) was then run from the last year of spin-up stage 3, forced by historical climate forcing and land cover maps, and rising CO

concentrations, as detailed below.

5.1.2 Atmospheric forcing datasets

The use of two different forcing datasets represents a first step in documenting atmospheric-forcing-based uncertainty in model output. Runoff has been shown for instance to be particularly affected by differences in precipitation from dif- ferent datasets (Fekete et al., 2004; Biancamaria et al., 2009), and by the methods to partition total precipitation volumes between rainfall and snowfall during the cold season (Had- deland et al., 2011). The bias of meteorological drivers also impacts the carbon budget (Zhao et al., 2012). A description of the two datasets used follows.

GSWP3 v0

This 3-hourly 0.5

global forcing product (1901–2007) was developed for the third phase of GSWP3 (http://hydro.iis.

u-tokyo.ac.jp/GSWP3/). It is based on the 20th Century Re- analysis (20CR) version 2 performed with the NCEP land–

atmosphere model (Compo et al., 2011). 20CR was dynami- cally downscaled to T248 (0.5

) resolution using the Global Spectral Model (GSM) by data assimilation using the spec- tral nudging technique (Yoshimura and Kanamitsu, 2008).

Bias corrections for precipitation, temperature and longwave and shortwave downward radiations were made using the GPCC v6 (Global Precipitation Climatology Centre), CRU TS v3.21 (Climate Research Unit), and SRB (Surface Radi- ation Budget) datasets, respectively. Precipitation was parti- tioned into rainfall and snowfall referring to the ratio of the downscaled 20CR, and wind-induced undercatch correction (Motoya et al., 2002) was applied separately. We upscaled the GSWP3 forcing for 1

spatial resolution.

CRUNCEP v7

This 6-hourly 0.5

global forcing product (1901–2015) is a combination of the annually updated CRU TS v3.24 monthly climate dataset (New et al., 2000) and NCEP re- analysis (Kalnay et al., 1996). The latter is only used to generate diurnal and daily anomalies added to CRU TS monthly means, after bi-linear interpolation to the 0.5

res- olution of CRU, except for the precipitations which were linearly interpolated. A threshold of 0

C in 2 m temper- ature was used to partition the precipitation into rainfall and snowfall in the CRUNCEP forcing. Rainfall, cloudi- ness, relative humidity and temperature are taken from the CRU, while the other fields (pressure, longwave radi- ation, windspeed) were directly derived from NCEP (see more details at https://vesg.ipsl.upmc.fr/thredds/fileServer/

store/p529viov/cruncep/readme.html). We upscaled the forc- ing to a 1

spatial resolution dataset, which can be found at https://vesg.ipsl.upmc.fr/thredds/catalog/store/p529viov/

cruncep/V7_1901_2015/catalog.html.

Table 1.

List of the datasets used for the ORCHIDEE-MICT evaluation, with their references, the original spatial resolution, and period of availability.

Dataset Variable Resolution Period URL References

Evaluation datasets for water budget

GRACE TWS 1^◦ Jul 2003–Dec 2007 http://grace.jpl.nasa.gov Swenson and Wahr (2006); Swenson (2012)

Landerer and Swenson, 2012

GlobSnow Snow water mass 25 km 1979–2013 www.globsnow.info Takala et al. (2009)

GLEAM v3.0a Evapotranspiration 0.25^◦ 1980–2014 http://www.gleam.eu Miralles et al. (2011)

GRDC River discharge – 1981–2007 http://www.bafg.de/GRDC/EN/Home/homepage_node.html –

Naturalized discharge River discharge – 1981–2007 http://www.r-arcticnet.sr.unh.edu/ObservedAndNaturalizedDischarge-Website Shiklomanov and Lammers (2009)

ESA CCI SM v2.2 Topsoil moisture 25 km Nov 1978–Dec 2014 http://esa-soilmoisture-cci.org –

GLEAM v3.0a Root-zone soil moisture 0.25^◦ 1980–2014 http://www.gleam.eu Martens et al. (2017)

Evaluation datasets for air-to-soil temperature continuum

ECA&D Snow depth – 1975–2005 http://ecad.knmi.nl/dailydata/predefinedseries.php –

National Climate Data and Snow depth – 1975–2005 http://climate.weather.gc.ca/historical_data/search_historic_data_e.html – Info. Archive of Env. Canada

USHCN Snow depth – 1975–2005 http://cdiac.ornl.gov/epubs/ndp/ushcn/ushcn.html –

RIHMI-WDC Snow depth – 1975–2005 – Bulygina et al. (2011)

National Meteo. Info. Snow depth – 1975–2005 – Peng et al. (2010)

Center of the China Meteo. Admin.

– Surface soil temperature 25 km 2000–2011 http://doi.pangaea.de/10.1594/PANGAEA.833409 André et al. (2015)

– In situ air and – 1980–2000 – Sherstiukov (2012)

soil temperatures

CALM Active-layer thickness – 1990–2015 – –

For Yakutia Active-layer thickness – 1960–1987 https://doi.org/10.1594/PANGAEA.808240 Beer et al. (2013)

Evaluation datasets for leaf area, carbon stocks and fluxes

GIMMS Leaf area index 0.08^◦ Jul 1981–Dec 2011 http://cliveg.bu.edu/modismisr/lai3g-fpar3g.html Zhu et al. (2013)

GLASS Leaf area index 0.05^◦ 1982–2012 http://glcf.umd.edu/data/lai Liang and Xiao (2012)

CAMS NEE 1.875^◦×3.75^◦ 1979–2015 https://apps-test.ecmwf.int/datasets/data/cams-ghg-inversions Chevallier et al. (2010)

Jena s96 v3.8 NEE 3.8^◦×5.0^◦ 1996–2015 http://www.bgc-jena.mpg.de/CarboScope/?ID=s96_v3.8 Rödenbeck (2005)

– GPP – Not known – Campioli et al. (2015)

MTE-GPP GPP 0.5^◦ 1982–2010 – Jung et al. (2009, 2011)

– NPP – Not known – Campioli et al. (2015)

MOD17A3.005 NPP 1 km 2000–2010 – –

NCSCD Soil carbon inventories 0.01^◦ – http://bolin.su.se/data/ncscd/netcdf.php Hugelius et al. (2013)

SoilGrids Soil carbon inventories 1 km – https://doi.org/10.5879/ecds/00000001 Hengl et al. (2014)

– Biomass carbon stocks 0.01^◦ – – Avitabile et al. (2016)

– Biomass carbon stocks 0.01^◦ – http://www.biomasar.orghttp://www.bgc-jena.mpg.de/geodb/projects/Home.php Thurner et al. (2014)

GFED4s Burned area 0.25^◦ 1997–2015 http://www.globalfiredata.org/data.html van der Werf et al. (2010)

and fire emissions

5.1.3 Vegetation and soil texture map

The ESA CCI Land Cover map (Bontemps et al., 2013) was used to produce the PFT map for ORCHIDEE. The ESA CCI land cover product is given by three maps at a 300 m spa- tial resolution, corresponding to the years 2010, 2005 and 2000. These maps were derived from the interpretation of MERIS full and reduced resolutions and SPOT-Vegetation time series. Land cover was classified according to the 22 classes used in the UN-LCCS (land cover classification sys- tem) scheme, which was translated into PFT fractions used in ORCHIDEE, following the cross-walking method presented by Poulter et al. (2011, 2015). Historical land use maps from the Harmonized Global Land Use dataset (Chini et al., 2014) were incorporated to reconstruct PFT fractions since 1860, following Peng et al. (2017), which will be detailed in the upcoming ORCHIDEE-TRUNK paper for CMIP6.

For soil texture, we use the 12 USDA texture classes pro- vided at a global 0.08

resolution from Reynolds et al. (2000) and upscaled these to the resolution of the atmospheric dataset (1

× 1

). Only the dominant texture type for a grid cell is used at the 1

resolution for defining soil hydraulic pa- rameters (Carsel and Parrish, 1988) in ORCHIDEE-MICT.

5.2 Evaluation datasets

The selected datasets used to evaluate ORCHIDEE-MICT are summarized in Table 1 and described in the Appendix.

For the water budget evaluation, we selected six Arctic river basins (Fig. 2b) which are important contributors to total Arctic Ocean river inflow: the four largest Eurasian Arctic basins (Ob, Yenisei, Lena and Kolyma), the Mackenzie Basin in northwestern Canada and the Yukon Basin in Alaska. The four Eurasian basins (with the smaller Pechora and Sever- naya Dvina basins) drain about two-thirds of the Eurasian Arctic landmass (Peterson et al., 2002), while the Macken- zie is the largest North American river, bringing freshwater to the Arctic Ocean (Woo and Thorne, 2003). We also eval- uated the Volga Basin (Fig. 2b), which is subject to snowfall events during the year but is not underlain by permafrost, in order to compare results with the high-permafrost Arctic basins (Fig. 2a and Table S4).

6 Evaluation of the atmosphere–snow–soil continuum

In the following, we analyze model performance in replicat-

ing the transfer of heat from atmosphere to deep soils. This is

done by evaluation against measured temperature gradients,

starting from snow depth controls on winter 1T , followed by

evaluation of surface (skin) temperature in summer, and the

M. Guimberteau et al.: ORCHIDEE-MICT, a LSM for the high latitudes 129

Figure 2. (a)

The three high-latitude sub-regions used in this study, including boreal North America (BONA), boreal Europe (BOEU) and boreal Asia (BOAS), following McGuire et al. (2016). Blue and red lines indicate the extent of continuous permafrost and all permafrost categories, respectively, according to the IPA permafrost map (Brown et al., 2002).

The seven high-latitude basins selected for this study with the gauge stations (red circles on the map, more information in Table S4).

temperature gradients between the near-surface and deeper soils.

6.1 Snow insulation controls on the temperature gradient between air and topsoil

6.1.1 Snow depth

ORCHIDEE-MICT correctly captures the spatial distribu- tion of maximum monthly average snow depth (Fig. 3a, b), and the seasonal decrease in snow depth from March to June (Fig. 3c), but modeled snow depth strongly depends on the atmospheric forcing used. GSWP3 climate forcing tends to produce a larger maximum snow depth than CRUNCEP, greater than those observed in all northern regions, especially over boreal Europe (BOEU) (Fig. 3c). This shows that uncer- tainty from climate forcing data is as large as the model bias compared with observations, making it difficult to attribute a model bias to a particular component of the snow model.

However, the rate of sublimation in winter (Pomeroy et al., 1998) and the prescribed albedo value of fresh snow have been shown to be critical in determining the peak value and phase of both snow depth and SWE (T. Wang et al., 2015).

6.1.2 Snow conductivity and snow density

Mean snow density and mean snow thermal conductiv- ity are computed at the month of maximum snow depth over the 1981–2007 period as weighted averages over the three snow layers. Gouttevin et al. (2012b) report density values of 200 kg m

for taïga and 330 kg m

for tun- dra and conductivity values of 70 mW m

K

for taïga and 250 mW m

K

for tundra, from Sturm and John- son (1992) and Domine et al. (2010). These higher values

over tundra were attributed to snow compaction by wind.

This process is not modeled in ORCHIDEE-MICT and we thus simulate similarly high values of conductivity for both biomes (Fig. S3): approximating tundra with C3 grass PFT between 55 and 85

N, and taïga with the boreal forests PFTs between 45 and 70

N and considering only grid cells with a fraction of the dominant biome above 0.6. The model yields a mean snow conductivity of 266 ± 203 (GSWP3) and 219 ± 197 (CRUNCEP) mW m

K

for tundra com- pared to 221 ± 113 (GSWP3) and 182 ± 100 (CRUNCEP) mW m

K

for taïga and a mean density of 269 ± 102 (GSWP3) and 239 ± 103 (CRUNCEP) kg m

for tundra and of 233 ± 67 (GSWP3) and 207 ± 63 (CRUNCEP) kg m

for taïga. Note that a recent study (Domine et al., 2016) sug- gests for tundra a complex structure with depth-hoar devel- oping at the base of snowpack during the course of the snow season, causing conductivities as low as 20 mW m

K

in late winter, whereas snow-compacted upper layers have con- ductivities of 200 to 300 mW m

K

, more comparable to ORCHIDEE-MICT.

6.2 Summer land surface temperature

ORCHIDEE-MICT overestimates summer (June–August)

LST by about 1.36

C when forced by GSWP3 (Fig. 4b)

and 0.49

C by CRUNCEP (Fig. 4c). The bias is larger in

northern and southern Siberia, where it can reach 4

C. These

differences may be linked to the overall underestimation of

ET (see Fig. 11). This is further addressed in the Discussion

(Sect. 10).

Figure 3.

Maximum monthly snow depth (m) simulated (background maps) with

GSWP3 and

CRUNCEP forcings compared to observations (color filled circles), averaged over the period 1975–2005.

Monthly mean seasonal snow depth (m) from observation and the two simulations, averaged over the observation sites in the three high-latitude sub-regions (shown in Fig. 2a).

6.3 Soil temperature

The simulated spatial patterns of mean annual topsoil (0.2 m) temperature generally reproduce the observed gradient along a southwest–northeast transect in Siberia (Fig. 5a, b). How- ever, the CRUNCEP-forced simulation results are colder than those from the GSWP3 ones as well as relative to observa- tions in the permafrost region (Fig. 5a, b), mainly driven by a strong cold bias in CRUNCEP-based winter soil tempera- tures (Fig. 5c, d).

During winter, the snowpack acts as an insulating layer above the soil surface, reducing soil heat loss. Snow thus causes a large positive temperature gradient (1T ), which is controlled by both snow depth and snow thermal proper- ties, such as thermal conductivity, density and albedo. Gen- erally, the model underestimates snow insulation in the early (November to January) and late (February to April) cold sea- sons for the same snow depth, as compared to observations (Fig. 6). This indicates that relatively congruent wintertime soil temperatures in the GSWP3-forced simulation in per- mafrost regions (Fig. 5c) may be due to a bias compensation from overestimated snow depth (Fig. 3) and underestimated snow thermal insulation.

A prominent component of the 1T –snow depth relation- ship observed at Russian stations (black in Fig. 6) is the sig-

nificantly lower insulation during the late cold season, prob- ably due to snow compaction and densification leading to higher snow conductivity (Decharme et al., 2016). This dif- ferential sensitivity of 1T to snow depth between the two pe- riods is effectively captured by ORCHIDEE-MICT, despite small modeled negative 1T values (i.e., topsoil colder than air) compared to observations at the termination of the snow season, when snow depth is diminished (< 20 cm).

Summer soil temperatures are higher in the GSWP3-

forced simulation relative to those of CRUNCEP, and warmer

than observations from the Russian meteorological stations

in continuous permafrost regions by 1 ∼ 2

C on average at

0.8 and 1.6 m depths (Fig. 5c, d). Spatially, however, the bias

in peak summer soil temperature varies for different regions,

with a large warm bias for the Lena Basin below 0.8 m, and

some cold bias for the further eastern sites (Fig. S4). This

is consistent with the overestimation of ALT for Yakutia

(Fig. S5) compared to the empirically derived map by Beer

et al. (2013) (see Sect. 6.4). Differences between the two

simulations in summertime soil temperatures may be partly

driven by warmer GSWP3 land skin temperatures during

summer (Fig. 4) than CRUNCEP. In addition, the cold bias in

winter soil temperatures in the CRUNCEP-forced simulation

might be carried over to summer via soil thermal inertia. This

M. Guimberteau et al.: ORCHIDEE-MICT, a LSM for the high latitudes 131

Figure 4. (a)

Observed mean summer (JJA) land surface temperature (

C) and bias in the

GSWP3 and

CRUNCEP-forced simulations, averaged over the period 1996–2007.

would partly offset the warm bias in CRUNCEP land surface temperatures during summer (Fig. 4c).

6.4 Active layer thickness and permafrost area

Figure 7a, b show the simulated spatial pattern of ALT, as calculated from modeled soil temperatures using a linear interpolation between soil layers to locate the first depth that remains frozen (below 0

C) year-round. Compared with CALM observations, in which most of the sites have ALT < 1 m, the GSWP3-forced model generally overesti- mates ALT by more than 1 m (also see Fig. S5, which compares modeled ALT of Yakutia, eastern Siberia with the map of Beer et al., 2013), whereas CRUNCEP-forced output shows relatively better agreement with the observa- tions. Apart from the uncertainty induced by climate forc- ing, the model–data mismatch may also arise from scale dif- ferences for the organic carbon content that is used to cal- culate soil physical properties for each grid cell. As men- tioned in Sect. 4.2, the empirical SOC map from NCSCD (Hugelius et al., 2013) is prescribed for permafrost regions in the soil thermal and hydrological modules, which is up- scaled by the model to the target spatial resolution (1

by 1

in this study). These SOC values thus do not represent

the site-level soil conditions, aside from the uncertainty of the NCSCD database itself. To further investigate this im- pact, we conducted additional simulations for the sites that provide explicit organic layer thicknesses (in total, 69 sites), forced by CRUNCEP. In these runs, we assumed pure or- ganic soil, i.e., f

in Eq. (9) equaling one, for the soil layers above the site-specific organic layer thickness, but kept the SOC concentration unchanged below this thickness, i.e., from NCSCD. Note that the moss layer, vegetation mat, and/or organic root zone as described in some sites were all summed to derive a total organic layer thickness. The other configurations including climate forcing and soil tex- ture were the same as the regional simulation. The result is displayed in Fig. S6, showing significantly shallower ALTs simulated by these site runs which better match the observa- tions (Fig. S6a), with different magnitudes of ALT reductions among the sites (Fig. S6b).

For simulated permafrost extent, two typical definitions of

permafrost in LSMs, one defined as ALT less than 3 m to

give “near surface” permafrost (e.g., Koven et al., 2013), the

other defined if any of the soil layers stay frozen (e.g., Ekici

et al., 2014; Burke et al., 2017b), produce quite different per-

mafrost extents (Fig. 7c, d). This result highlights that the

intercomparison of permafrost areas among different LSMs

Figure 5.

Mean annual soil temperature at 0.2 m depth (

C) in the

GSWP3 and

CRUNCEP-forced simulations (background maps), compared to the site observations (color filled circles), averaged over the period 1981–2000. Monthly mean seasonal soil temperatures at different depths (

C) in the

GSWP3 and

CRUNCEP-forced simulations, compared with the observation, averaged over the 51 sites in continuous permafrost region (according to the IPA map) and over the period 1981–2000. The spatial patterns of maximum monthly soil temperature are also shown in Fig. S4.

Figure 6.

Relationship between

(soil temperature at 20 cm depth minus air temperature) and snow depth (cm) over the period 1981–2000, for site-level observations (black), and for model results (red) (9612 site-month values in total), forced by

GSWP3 and

CRUNCEP.

Circles and squares are medians of 5 cm snow depth bins, representing the early (November–January) and late (February–April) snow season, respectively. Upper and lower bars indicate the 25th and 75th percentiles of each bin. The size of circles/squares indicates the frequency of occurrence in each bin.

with differing soil vertical discretizations may include uncer- tainties brought by the arbitrarily chosen definition, whereas evaluation and comparison directly for soil temperatures and ALT should be more robust. A qualitative comparison against the empirical IPA (International Permafrost Association) per- mafrost map (Brown et al., 2002) shows better agreement for CRUNCEP compared to GSWP3-forced output, since CRUNCEP-forced simulation generally matches the distri- bution of continuous permafrost using the former definition, while GSWP3-forced simulation seems to underestimate per-

mafrost extent (Fig. 7c, d). This is consistent with the deeper simulated ALT under GSWP3 climate forcing.

7 Evaluation of large-scale water storage and fluxes

Simulated water budget components are evaluated over se-

lected northern basins (Fig. 2b), most of which are underlain

by permafrost (e.g., Lena, Kolyma), with the exception of

the warmer Volga. The Ob and Yenisei catchments have con-

trasting north–south precipitation and temperature regimes,

M. Guimberteau et al.: ORCHIDEE-MICT, a LSM for the high latitudes 133

Figure 7.

Active layer thickness (ALT in m) from the

GSWP3 and

CRUNCEP-forced simulations (background maps), compared to the observed ALT from the CALM network (color filled circles), averaged over the period 1990–2007. Permafrost extent from the

GSWP3 and

CRUNCEP-forced simulations according to two different definitions (yellow and red lines) on top of the IPA permafrost map (Brown et al., 2002).

Figure 8.

Interannual monthly variation and trend (line) of TWS (mm) simulated with the two forcings compared to GRACE data over the seven basins (see Fig. 2b), for the period July 2003–December 2007.

with attendant impacts on Arctic Ocean discharge and sub- basin-scale water budgets. Here, only basin-scale averages are discussed.

7.1 Total terrestrial water storage change

A realistic phase and amplitude of TWS are simulated with both forcing datasets, although peak-to-peak amplitude is

slightly overestimated in the Volga, Yenisei and Kolyma

basins under GSWP3 input and the seasonal amplitude un-

derestimated in the Yukon with both forcings, and in the

Mackenzie and Lena with CRUNCEP forcing (Fig. 8). The

positive temporal trend of TWS in the Ob, Lena and Kolyma

basins is captured well by ORCHIDEE-MICT, where it re-

flects upward precipitation trends (not shown). In general,

the GRACE TWS is captured well by ORCHIDEE-MICT

with both forcings, at the seasonal scale and for the 5-year trends, except in the Yukon Basin, where observed TWS de- creases are not reproduced in our simulations, with no pre- cipitation decrease in the GSWP3-forced model. The Yukon TWS decline is likely due to glacier melt in the northwestern Cordillera (S. Wang et al., 2015). As glaciers are not rep- resented in ORCHIDEE-MICT, the model does not capture these TWS trends. Note that groundwater storage changes re- lated to the development of closed and open taliks (Muskett and Romanovsky, 2009) which contribute to existing TWS trends – increasing storage in the Lena and Yenisei, decreas- ing it in the Mackenzie Basin, and no change in the Ob – are not modeled either in ORCHIDEE-MICT. Despite this, the model reproduces observed trends in these basins.

7.2 Snow-related processes controlling land water storage in the cold season

The modeled seasonal cycle of the SWE and the length of the snowmelt period are in agreement with observations (Fig. 9), suggesting a good parameterization of the snowmelt and sub- limation processes. However, results differ strongly accord- ing to forcing inputs. In basins with a large permafrost frac- tion (Yukon, Lena and Kolyma) and, to a lesser extent, in the Mackenzie, Ob and Yenisei basins, SWE is underesti- mated throughout the year compared to GlobSnow data when ORCHIDEE-MICT is forced by CRUNCEP, and it is signif- icantly larger in the GSWP3-forced simulation (in the Volga Basin, the SWE is overestimated in the two simulations).

This is due to the low basin-specific snowfall rate in CRUN- CEP forcing compared to GSWP3 (Fig. S12), which is prob- ably the result of the criterion used to partition rainfall and snowfall in CRUNCEP, and strongly affects the simulation of snow depth and SWE (Loth et al., 1993; Wen et al., 2013). By contrast, the GSWP3-forced model captures the early winter SWE accumulation in these basins. In spring, the SWE is sys- tematically overestimated except over the Lena, whose sea- sonal cycle is well reproduced by ORCHIDEE-MICT. This corresponds to an excessive persistence of the snow cover, which may be explained by the absence of hysteresis in the snow depletion curve relating the snow cover extent and the SWE (e.g., Magand et al., 2014). In the Yenisei and Macken- zie basins, the SWE in winter is closer to observations with the GSWP3 forcing, with the exception of springtime val- ues, which are better under CRUNCEP forcing. In basins where the permafrost area is near-zero (Ob and Volga), the SWE from CRUNCEP-forced simulation is closer to Glob- snow than those from GSWP3, in which SWE is overesti- mated in winter and spring. This is related to the large differ- ence in snowfall over these basins between the two forcings (Fig. S12).

7.3 Soil moisture In the topsoil

Seasonal evolution of topsoil moisture (first 2 cm) is com- pared to the ESA-CCI-SM product over the seven basins (Fig. 10a). Liquid soil moisture values from the model are used for the comparison because the ESA-CCI-SM prod- uct measures only the topsoil moisture when temperature is above 0

C. Because of scaling issues (the satellite product is rescaled to the 10 cm top layer of the NOAH model, as al- ready noted, but is more representative of a thinner soil layer of about 2 cm), the comparison was performed on relative liquid soil moisture values after normalization with their re- spective SD. The observed seasonal moisture variations are captured well by ORCHIDEE-MICT with both forcings over the seven basins (Fig. 10a). The maximum values occur in summer, in contrast to lower latitudes, because of the thaw- ing processes occurring in summer. The local minimum sim- ulated in summer in the Volga and Ob basins is underesti- mated, suggesting a too slow infiltration front of the water in the topsoil layers of the model. Thus, less water in the root zone is available for transpiration, which is found to be underestimated when compared to GLEAM (see Fig. 11b).

Compared to observations, a more rapid increase (decrease) in the modeled topsoil moisture in spring during snowmelt (in fall) is found in the Yukon and Mackenzie basins, related to a more rapid thawing (freezing) of the topsoil.

In the root zone

The soil water deficit is of primary importance during spring and summer at the high latitudes because of its potential impacts on the vegetation transpiration, leading to a sur- face temperature increase and a reduction in the productiv- ity. However, a soil water comparison between GLEAM and ORCHIDEE-MICT is difficult because of differences in soil depth, which in GLEAM varies with vegetation cover, but is fixed at 2 m in ORCHIDEE-MICT. Moreover, the fraction of soil tiles of short vegetation and forests used in the two prod- ucts is not the same. We therefore normalize the relative root soil moisture (Fig. 10b) by its SD to compare the dynamic of the soil moisture rather than the total amount of water in the soil. The intensity of water uptake by the roots of the veg- etation in summer is generally well simulated by the model in both simulations. In basins underlain by permafrost, ex- cept in the Lena Basin, water uptake is delayed by 1 month for both forcing sets, while the rate of decrease in root soil moisture is underestimated for GSWP3-forced output only.

The similarity in output in this respect, despite very different

SWE, highlights the low impact of the latter on the root soil

moisture, and further underscores how the snowmelt differ-

ential is lost through runoff rather than being available for

vegetation.

M. Guimberteau et al.: ORCHIDEE-MICT, a LSM for the high latitudes 135

Figure 9.

Monthly mean seasonal SWE (mm) simulated with the two forcings compared to GlobSnow over the seven basins (see Fig. 2b), averaged over the period 1981–2007.

Figure 10.

Monthly mean seasonal relative

topsoil (–) and

root soil moisture (–), both normalized by their multi-year SD, simulated with the GSWP3 and CRUNCEP forcings over the seven basins (see Fig. 2b), averaged over the period 1981–2007. The results are compared with

satellite-derived observations from ESA CCI and

the GLEAM data-driven model assimilated against satellite data.

7.4 Evapotranspiration and component fluxes

The amplitude of the ET seasonal cycle is generally well cap- tured by the model in most basins, when compared to the GLEAM product, despite a systematic underestimation by

ORCHIDEE-MICT, whatever the input forcing (Fig. 11a).

The disparity between modeled peak ET and GLEAM data

is reduced under GSWP3 forcing in the Yukon, and under

CRUNCEP forcing in the Yenisei and Lena. ET increase is

Figure 11. (a)

Monthly mean seasonal evapotranspiration (mm d

) simulated with the two forcings compared to the GLEAM data-driven model over the seven basins (see Fig. 2b), averaged over the period 1981–2007.

Seasonal bias of ET components (mm d

) averaged over the same period, with the GSWP3 (solid line) and CRUNCEP (dashed line) forcings.

underestimated in spring and early summer for both forcings, except in the Volga Basin. This is consistent with modeled LAI increasing too late in spring (see Fig. 13), which could be due, at least partially, to an excessive persistence of the snow cover in spring. In fall, the timing of the decrease in ET is reproduced by both forcings. Model biases with re- spect to GLEAM data in the sublimation, soil evaporation and transpiration components of ET are shown in Fig. 11b. In all basins, sublimation bias in simulations forced by GSWP3 is ∼ 0 in winter, in agreement with GLEAM. By contrast, CRUNCEP-forced simulations slightly overestimate subli- mation in early spring across basins with the exception of the Volga. These results are consistent with SWE underesti- mation (except in the Volga) (Fig. 9) and higher downward shortwave radiation (Fig. S14) that results when CRUNCEP forcing is used. The general underestimation of summer ET by ORCHIDEE-MICT in the Yukon, Mackenzie and Kolyma basins is explained mainly by a too low transpiration despite bare soil evaporation being slightly overestimated (Fig. 11b).

When forced by CRUNCEP data, ORCHIDEE-MICT over- estimates interception loss in all basins but the Kolyma and Yukon, which is consistent with CRUNCEP LAI overestima- tion (see Fig. 13).

7.5 River discharge

By comparing the two simulations, it is clear that the mete- orological forcing exerts a significant influence on the sim- ulated river discharge (Fig. 12): GSWP3 leads to systemat- ically higher river discharge than CRUNCEP, which is per- fectly consistent with the SWE biases (Fig. 9). In a ma- jority of basins, GSWP3-forced simulations better capture the seasonal cycle of observed discharge than CRUNCEP- forced ones, especially in the Yukon and eastern Siberian basins, where the discharge is strongly underestimated under CRUNCEP forcing (between 60 % in the Yenisei and 83 % in the Yukon).

A first feature of the nival regime characterizing the stud-

ied high-latitude basins is the occurrence of low flows in win-

ter, when water is frozen in the snowpack, soils, and river

ice. This is well simulated by ORCHIDEE-MICT, despite

a small underestimation in the Yukon, Mackenzie and Yeni-

sei. Naturalized river discharge is available in the latter, and

lower in winter than GRDC values, which reflects the effect

of reservoir operations. The winter discharge simulated by

ORCHIDEE-MICT in the Yenisei is closer to the naturalized

estimates, as the model does not account for artificial reser-